Intelligent parameter calibration method and device for diesel engine using improved grey wolf algorithm in digital twin model

Abstract

The improved grey wolf algorithm is used in a diesel engine intelligent parameter calibration method and device in a digital twin model, and relates to the field of diesel engine combustion performance analysis. To solve the technical problem in the prior art that, in the initialization process of the traditional grey wolf algorithm, only random selection is performed, although the calculation resources are less called, the convergence is easier, but the local optimum is easily fallen into, the technical scheme provided by the present application is that: the improved grey wolf algorithm is used in a diesel engine intelligent parameter calibration method in a digital twin model, the method comprising: collecting diesel engine cylinder parameters and preprocessing to obtain at least three points as input quantities of the grey wolf algorithm; collecting preset input quantities, a target function and a stop iteration condition of a pre-calibration model; solving the initial iteration point of the grey wolf algorithm, and starting iteration to eliminate weak grey wolves until the preset stop iteration condition is reached; obtaining the grey wolf position according to the current grey wolf algorithm as the calibration result. It is suitable for application in intelligent calibration of diesel engine model parameters.

Description

TECHNICAL FIELD

[0001] The application relates to the field of diesel engine combustion performance analysis, in particular to an intelligent calibration method for diesel engine model parameters. BACKGROUND

[0002] As a key power device of the main engine of a ship, the diesel engine undergoes a series of energy conversions, and converts the chemical energy of diesel fuel into heat energy through combustion. Then, the heat energy is further converted into mechanical energy through expansion to output work. In view of the escalating restrictions on environmental and energy policies, optimizing the operating conditions of the diesel engine to meet the green, energy-saving and safety requirements has always been the most concerned problem of researchers in the field. With the deep integration of cutting-edge technologies such as artificial intelligence, Internet of Things (IoT) and big data, the shipbuilding industry is particularly facing unprecedented challenges and opportunities. With the development of autonomous ships, the exploration of intelligent ships has become a key research direction in the industry. Under the background of intelligent development, digital twinning technology is attracting more and more attention, and the core is to integrate multi-physical and multi-scale models with high fidelity. Based on these high-precision models, dynamic and interactive data streams are used to draw comprehensive life cycle activities of objects in the physical and virtual fields, and finally promote the realization of intelligent development goals. Digital twinning technology has great prospects because it supports the deep integration of the real world and the virtual world. Although the application of digital twinning technology in the field of marine diesel engines is still limited at present, the prospect it provides has great value, which highlights the necessity of a comprehensive research initiative.

[0003] Digital twin technology combines high-fidelity models with real-world data as a means to support the development of various system intelligence, which has attracted attention. As a promising means to promote seamless integration of the real world and the virtual world, although the application of digital twin technology in marine diesel engines is still limited, its future application potential has great value and is worth comprehensive research. Digital twin technology has the advantages of reducing cost, reducing risk, improving efficiency, and real-time monitoring and predicting engine performance. However, digital twin diesel engine modeling involves many parameters that are difficult to measure directly, resulting in lower accuracy of experimental data. Therefore, it is necessary to calibrate multiple parameters simultaneously to achieve accurate calibration of the digital twin model. As the complexity of the diesel engine model increases, the amount of calibration work increases exponentially, and manual calibration becomes challenging. To address these challenges, the grey wolf algorithm can be used for fast and intelligent calibration of digital twin models. The grey wolf algorithm has the advantages of simplicity, exploration and development balance, simple parameter algorithm involved in the use process, and faster convergence speed. However, in the initialization process of the traditional grey wolf algorithm, only random selection is performed, which although reduces the computational resources called and is easier to converge, the initial random placement of the grey wolf can significantly affect the search process. In many cases, multiple runs with different initializations are required to obtain satisfactory results, and the algorithm has poor robustness, different random inputs can cause significant differences in optimization results, and can easily fall into local optima. SUMMARY

[0004] To solve the technical problems in the prior art that the traditional grey wolf algorithm only randomly selects in the initialization process, which although reduces the computational resources called and is easier to converge, the initial random placement of the grey wolf can significantly affect the search process, and in many cases, multiple runs with different initializations are required to obtain satisfactory results, and the algorithm has poor robustness, different random inputs can cause significant differences in optimization results, and can easily fall into local optima, the technical solution provided by the present application is:

[0005] The diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model, the method comprises:

[0006] The step of collecting diesel engine cylinder parameters and preprocessing to obtain at least three points as input quantities of the grey wolf algorithm;

[0007] The step of collecting the preset input quantity, objective function and stopping iteration condition of the pre-calibration model;

[0008] The step of solving the initial iteration point of the grey wolf algorithm and starting iteration to eliminate weak grey wolves until the preset stopping iteration condition is reached;

[0009] The step of obtaining the grey wolf position according to the current grey wolf algorithm as the calibration result.

[0010] Further, a preferred embodiment is provided, wherein the preprocessing comprises:

[0011] The diesel engine cylinder parameters collected are processed using three-point quadratic interpolation, so that the interpolation difference between the parameters and the simulation parameters is not greater than a preset value.

[0012] Further, a preferred embodiment is provided, wherein the diesel engine cylinder parameters include but are not limited to cylinder pressure frequency and crank angle.

[0013] Further, a preferred embodiment is provided, wherein the initial iteration point is solved by a Monte Carlo algorithm, and the starting iteration is implemented by a greedy algorithm.

[0014] Further, a preferred embodiment is provided, wherein the objective function is specifically:

[0015]

[0016] wherein N is the total number of discrete points to be compared, j represents the number, j = 1, 2, 3, …, represents the jth model simulation data, represents the jth experimental data.

[0017] Further, a preferred embodiment is provided, wherein in the iteration process, the first three best solutions obtained at each iteration are saved, and the performance of other grey wolf individuals is improved by updating the position of the population.

[0018] Based on the same inventive concept, the present application also provides a diesel engine intelligent parameter calibration device using an improved grey wolf algorithm in a digital twin model, which comprises:

[0019] A module for collecting diesel engine cylinder parameters and preprocessing to obtain at least three points as input quantities of the grey wolf algorithm;

[0020] A module for collecting preset input quantities, an objective function and a stopping iteration condition of a pre-calibration model;

[0021] A module for solving the initial iteration point of the grey wolf algorithm and starting iteration to eliminate weak grey wolves until a preset stopping iteration condition is reached;

[0022] A module for obtaining the position of the grey wolf according to the current grey wolf algorithm as a calibration result.

[0023] Based on the same inventive concept, the present application also provides a digital twin diesel engine model, which is calibrated by the method.

[0024] Based on the same inventive concept, the present application also provides a computer storage medium for storing a computer program, which, when read by a computer, causes the computer to perform the method.

[0025] Based on the same inventive concept, the present application also provides a computer comprising a processor and a storage medium, which, when read by the processor, causes the computer to perform the method.

[0026] Compared with the prior art, the technical solution provided by the present application has the following advantages:

[0027] The diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by the present application adopts the grey wolf algorithm to complete intelligent calibration of model parameters, thereby shortening the model development time.

[0028] The diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by the present application improves the traditional grey wolf algorithm.

[0029] The diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by the present application is suitable for intelligent calibration of diesel engine model parameters. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 The flowchart of the improved grey wolf algorithm for the first embodiment is shown.

[0031] Figure 2 The flowchart of the diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by the first embodiment is shown. DETAILED DESCRIPTION

[0032] In order to make the advantages and benefits of the technical solutions provided by the present application more clear, the technical solutions provided by the present application will be further described in detail in conjunction with the drawings, specifically:

[0033] Embodiment one, in conjunction with Figure 1 and 2 The present embodiment provides a diesel engine intelligent parameter calibration method using an improved grey wolf algorithm in a digital twin model, which comprises the following steps:

[0034] Collecting diesel engine cylinder parameters and preprocessing to obtain at least three points as input quantities of the grey wolf algorithm.

[0035] collecting preset input quantity, objective function and stopping iteration condition of the pre-calibration model;

[0036] solving initial iteration point of the grey wolf algorithm and starting iteration to eliminate weak grey wolves until preset stopping iteration condition is reached;

[0037] obtaining grey wolf position according to current grey wolf algorithm as calibration result.

[0038] The three points are: collecting diesel engine cylinder parameters and preprocessing, obtaining diesel engine cylinder pressure curve, and discretizing the cylinder pressure curve into at least three points.

[0039] Specifically,

[0040] comprising the following steps:

[0041] 1. Data preprocessing: the frequency of collecting cylinder pressure by the experimental sensor is δ exp , and the collection of all cylinder pressure data is The corresponding crank angles are m is the number of points of the collected cylinder pressure, wherein m = 1, 2, 3…, and when constructing the diesel engine model oriented to digital twinning, the cylinder pressure data output by the model simulation is The corresponding crank angles are n is the number of cylinder pressure points output by the model simulation, n = 1, 2, 3…; wherein f is a function, and the function is used to indicate that this algorithm can calibrate any parameter that needs to be checked. If the heat dissipation rate is checked, f is the heat release rate. If the cylinder pressure is checked, f is the cylinder pressure. If other parameters are checked, f is other parameters; the experimental cylinder pressure curve is discretized into I points; (is the output of the diesel engine model, and is the input of the particle swarm algorithm) the cylinder pressure corresponding to each crank angle is If the sampling frequency of the experimental data does not match the crank angle interval of the simulation output, the cylinder pressure at the crank angle is approximated by using three-point quadratic Newton interpolation, and the calculation is performed in the form of difference quotient:

[0042] The second-order difference quotient formula and the third-order difference quotient formula are as follows:

[0043]

[0044]

[0045] wherein, represents the second-order difference quotient of the two adjacent points, represents the third-order difference quotient of the three adjacent points, The function value corresponding to the crank angle is generally the cylinder pressure collected by experiment, and can also be the heat release rate, gas flow energy, cylinder wall heat dissipation rate, etc. The embodiment mainly analyzes data processing and model calibration, and is not directed to a specific diesel engine model. In theory, the output of any diesel engine model can be applied.

[0046] The three-point quadratic Newton interpolation formula is:

[0047]

[0048] Step 2: Set the objective function

[0049] In the traditional grey wolf algorithm, the objective function is a function with a certain mathematical expression. In order to apply the grey wolf algorithm to the intelligent calibration of the diesel engine model facing digital twinning, the objective function is set as follows:

[0050]

[0051] Implementation;

[0052] Wherein, N is the total number of discrete points to be compared, j represents the number, j = 1, 2, 3,... The jth model simulation data is represented by j. The jth experimental data is represented by j.

[0053] Step 3: Pre-set initial stop iteration condition, which generally meets any of the following conditions:

[0054] 1. The number of remaining grey wolf populations is less than ε n ;

[0055] 2. The arithmetic mean of the objective function of the optimal grey wolf population and the arithmetic mean of the objective function of the worst grey wolf population in the continuous two initialization iteration processes satisfy the following formula:

[0056]

[0057] Wherein, Y ini_Best is the arithmetic mean of the objective function corresponding to the optimal grey wolf population, Y ini_Worst is the arithmetic mean of the objective function corresponding to the worst grey wolf population, ε n is a positive integer defined by the user, which is usually 4-10, and ε is a positive number greater than 0 and less than 1, which is usually 0.1-0.3.

[0058] Step 4: Grey wolf algorithm initialization

[0059] Monte Carlo method and greedy block method are combined to obtain the initial solution. N grey wolves are selected, if there are k parameters to be calibrated, each grey wolf carries k-dimensional information, each dimension is divided into 2k For each population, calculate the arithmetic mean of the objective function for different Monte Carlo plots, and eliminate 2. k-1 Weak gray wolf populations with larger arithmetic mean values ​​for the objective functions are retained, while the remaining superior gray wolf populations are preserved. When the initial iteration conditions are met, iteration stops, initialization ends, and the objective function value corresponding to the current optimal calibration parameters is stored as Y. ini_best_temp .

[0060] Step 5: The Grey Wolf algorithm is used to calibrate the parameters of the diesel engine model.

[0061]

[0062]

[0063] Where t represents the current iteration number. and It is a coefficient vector. As an intermediate variable, It is the position vector of the prey, that is Let t represent the position vector of the gray wolf omega, and t represent the current generation selection number.

[0064] vector and The calculation is as follows:

[0065]

[0066]

[0067] in The component decreases linearly from 2 to 0 during the iteration process. It is a random vector in [0,1].

[0068] The formula for updating the position of the three gray wolves in the gray wolf algorithm is as follows:

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] Wherein the subscripts 1, 2, 3 represent the positions of the updated alpha wolf, beta wolf and delta wolf respectively, and the subscripts a, b, d represent the positions of the alpha wolf, beta wolf and delta wolf at the last iteration respectively, are all intermediate coefficients.

[0077] In the grey wolf algorithm, the first three best solutions obtained so far are saved, and other grey wolf individuals, i.e. search agents (including omega), are forced to update their positions according to the best positions.

[0078] The grey wolf attacks the prey when the prey stops moving, thus completing the hunt. In order to mathematically model the approach to the prey, the value of is reduced during the iteration process; The fluctuation range of is also reduced, and |A|<1 forces the grey wolf to attack the prey, and the parameter values of the diesel engine model and the objective function value are output at the end of the iteration.

[0079] The grey wolf algorithm simulates the leadership hierarchy and hunting mechanism of grey wolves in nature; four types of grey wolves-alpha, beta, delta and omega are used to simulate the leadership hierarchy. The grey wolf algorithm can provide very competitive results in the optimization process, especially in the engineering field.

[0080] Embodiment two, the embodiment is a further limitation of the diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by embodiment one, and the preprocessing includes:

[0081] Using three-point quadratic interpolation to process the collected diesel engine cylinder parameters, so that the difference between the parameters and the interpolated parameters obtained by simulation is not greater than a preset value.

[0082] Embodiment three, the embodiment is a further limitation of the diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by embodiment one, and the diesel engine cylinder parameters include: cylinder pressure frequency and crank angle.

[0083] Embodiment four, the embodiment is a further limitation of the diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by embodiment one, and the initial iteration point is solved by the Monte Carlo algorithm, and the greedy algorithm is used to realize the start iteration.

[0084] Embodiment five, the embodiment is a further limitation of the diesel engine intelligent parameter calibration method using the improved grey wolf algorithm in the digital twin model provided by embodiment one, and the objective function is specifically:

[0085]

[0086] wherein N is the total number of discrete points to be compared, j represents the number, j = 1, 2, 3, …, represents the jth model simulation data, represents the jth experimental data.

[0087] Embodiment six, this embodiment is a further limitation of the diesel engine intelligent parameter calibration method using improved grey wolf algorithm in the digital twin model provided by embodiment one, in the iteration process, the first three optimal solutions obtained at each iteration are saved, and the performance of other grey wolf individuals is improved by updating the position of the population.

[0088] Embodiment seven, this embodiment provides a diesel engine intelligent parameter calibration device using improved grey wolf algorithm in a digital twin model, the device comprises:

[0089] A module for collecting diesel engine cylinder parameters and preprocessing to obtain at least three points as input quantities of the grey wolf algorithm;

[0090] A module for collecting preset input quantities, objective functions and stopping iteration conditions of the pre-calibration model;

[0091] A module for solving the initial iteration point of the grey wolf algorithm and starting iteration to eliminate weak grey wolves until the preset stopping iteration condition is reached;

[0092] A module for obtaining the position of the grey wolf according to the current grey wolf algorithm as the calibration result.

[0093] Embodiment eight, this embodiment provides a digital twin diesel engine model, which is calibrated by the method provided in embodiment one.

[0094] Embodiment nine, this embodiment provides a computer storage medium for storing a computer program, when the computer program is read by a computer, the computer executes the method provided in any one of embodiments one to six.

[0095] Embodiment ten, this embodiment provides a computer comprising a processor and a storage medium, when the processor reads the computer program stored in the storage medium, the computer executes the method provided in any one of embodiments one to six.

[0096] The above describes the technical solutions provided by the present application in further detail through several specific embodiments, in order to highlight the advantages and benefits of the technical solutions provided by the present application, but the above several specific embodiments are not used as a limitation of the present application, any reasonable modification and improvement of the present application, combination and equivalent replacement of embodiments, etc. based on the spirit and principles of the present application, should be included in the protection scope of the present application.

[0097] In the description of the present specification, only the preferred embodiments of the present application are described, and the present application is not limited by the preferred embodiments; in addition, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms is not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. Furthermore, the skilled person in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction. In addition, the terms "first", "second" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include at least one feature. In the description of the present application, the meaning of "N" is at least two, for example, two, three, etc., unless otherwise specifically limited. Any process or method described in the flowchart or otherwise described herein can be understood as a module, segment or part of code including executable instructions for implementing custom logic functions or processes of one or more steps, and the scope of the preferred embodiments of the present application includes additional implementations in which the functions can be performed in the order shown or discussed, including in a substantially simultaneous manner or in reverse order according to the functions involved, which should be understood by those skilled in the art. The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a list of executable instructions for implementing logical functions, which can be embodied in any computer readable medium for use by an instruction execution system, device or apparatus, such as a computer based system, a system including a processor or other system that can fetch and execute instructions from the instruction execution system, device or apparatus. For the present specification, "computer readable medium" can be any device that can contain, store, communicate, propagate or transport programs for use by an instruction execution system, device or apparatus or in conjunction with these instruction execution systems, devices or apparatus. More specific examples (non-exhaustive list) of computer readable medium include the following: electrical connections having one or N wires (electronic devices), portable computer disk cartridges (magnetic devices), random access memories (RAM), read only memories (ROM), erasable programmable read only memories (EPROM or flash memory), fiber optic devices, and portable compact disk read only memories (CDROM).Additionally, a computer readable medium can be paper or other comparable effectively medium whereupon at least one of the programs is printed, since the programs can be electronically retrieved, for instance by optically scanning the paper or other medium, then rendered into an electronically useable form by processes such as optical character recognition, and stored for access by a computer. Note that the foregoing approach has been described in relation to discrete embodiments, and analogous processes involving software or firmware instructions used in the setting of one or more appropriately programmed computers can also be implemented and / or utilized. Furthermore, the various embodiments can be implemented in hardware, software, or firmware, or in any combination thereof. In the above embodiments, the steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following technique, known in the art, or their combination, can be used: discrete logic circuitry having logic gates for implementing logic functions upon data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and the like.

[0098] Those skilled in the art can understand that all or part of the steps of the method carried out by the above-mentioned embodiments can be completed by programs instructing relevant hardware, and the programs can be stored in a computer readable storage medium. When the programs are executed, one or a combination of the steps of the method embodiments is included. In addition, each functional unit in each embodiment of the present application can be integrated into one processing module, or each unit can exist physically independently, or two or more units can be integrated into one module. The integrated module can be realized in the form of hardware or in the form of a software function module. If the integrated module is realized in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

Claims

1. A diesel engine intelligent parameter calibration method using an improved grey wolf algorithm in a digital twin model, characterized in that, The method comprises: a step of collecting diesel engine cylinder parameters and preprocessing to obtain at least three points as input quantities of the grey wolf algorithm; a step of collecting preset input quantities, an objective function and a stop iteration condition of a pre-calibration model; a step of solving initial iteration points of the grey wolf algorithm and starting iteration to eliminate weak grey wolves until a preset stop iteration condition is reached; a step of obtaining a grey wolf position according to a current grey wolf algorithm as a calibration result; the initial iteration points are solved by a Monte Carlo algorithm, and the starting iteration is realized by a greedy algorithm; in the iteration process, the first three best solutions obtained at each iteration are saved, and the performance of other grey wolf individuals is improved by updating the position of the group.

2. The method according to claim 1, wherein the improved grey wolf optimization algorithm is used in the intelligent parameter calibration of the diesel engine in the digital twin model. The preprocessing comprises: a step of processing the collected diesel engine cylinder parameters by using three-point quadratic interpolation, so that the interpolation difference between the parameters and the parameters obtained by simulation is not greater than a preset value.

3. The method of intelligent parameter calibration of diesel engine using improved grey wolf algorithm in digital twin model according to claim 1, characterized in that, The diesel engine cylinder parameters include but are not limited to cylinder pressure frequency and crank angle.

4. The method of intelligent parameter calibration for diesel engine using improved grey wolf algorithm in digital twin model according to claim 1, characterized in that, The objective function is specifically: ， wherein, is the total number of discrete points for which contrast is desired, denotes the number of, , denotes the simulation data of the th model, denotes the experimental data of the th experiment.

5. A diesel engine intelligent parameter calibration device using an improved grey wolf algorithm in a digital twin model, characterized in that, The device for realizing the method of claim 1 comprises: a module for collecting diesel engine cylinder parameters and preprocessing to obtain at least three points as input quantities of the grey wolf algorithm; a module for collecting preset input quantities, an objective function and a stop iteration condition of a pre-calibration model; a module for solving initial iteration points of the grey wolf algorithm and starting iteration to eliminate weak grey wolves until a preset stop iteration condition is reached; a module for obtaining a grey wolf position according to a current grey wolf algorithm as a calibration result.

6. A digital twin diesel engine model, characterized by, The model is calibrated by the method of claim 1.

7. Computer storage medium for storing a computer program, characterized in that When the computer program is read by the computer, the computer executes the method of any one of claims 1-4.

8. A computer comprising a processor and a storage medium, characterized in that When the processor reads the computer program stored in the storage medium, the computer executes the method of any one of claims 1-4.

Images